The goal of this proposal is to create a mathematical model of folate-mediated one-carbon metabolism (FOCM) and its relation to DNA methylation. This model will integrate knowledge of enzyme kinetics, genetics and epigenetics, and nutrition, and will enable us to investigate (1) mechanisms by which dietary factors influence DNA methylation, and (2) increase our understanding of these processes in cancer prevention. Altered folate metabolism is associated with changes in global and regional DNA methylation patterns, deficiencies in DNA synthesis and repair, and altered methionine cycle kinetics. Each of these factors may contribute to the increased cancer risk seen among individuals with a low folate status. FOCM is highly complex, with genetic (i.e., polymorphisms in folate-dependent enzymes) and nutritional influences (i.e., intakes of folate, vitamins B2, B6, B12, methionine, choline, and alcohol) interacting in intricate, but insufficiently understood, ways. Mathematical modeling is an approach that has been particularly useful in the study of such complex, non-linear biological systems. We propose to develop a mathematical model of FOCM and methylation, based on a modular approach, with extensive empirical testing of the model using data from experimental studies. Subsequently, we will investigate the impact of combined nutritional and genetic variation in FOCM on relevant biomarkers (methylation capacity, including DNA methylation; plasma and intracellular levels of S-adenosylmethionine and S-adenosylhomocysteine, red blood cell folate; homocysteine concentrations; nucleotide synthesis). We will investigate whether the mathematical model of FOCM is consistent with results from epidemiologic studies of folate biomarkers and colorectal neoplasia and is able to provide new insights into biological mechanisms. Finally, we propose to use simulation techniques within this model for identifying those components of FOCM that are most sensitive or robust to variation in nutritional intake and genetic alterations in enzyme activity. Our preliminary results with a model for the methionine cycle indicate that these goals are achievable. We plan to make the model generally available to the research community, and anticipate that it will become a useful tool that can be used as an adjunct to experimental investigations and as an aid to study design.